Comparing Output from Hierarchial Clustering Vs K Means

[mess2010]

We've been set a task to use Cluster analyse a large quantity of data using PASW(SPSS). The process we followed was:
1) Run Cluster anaylsis (using Hierarchial Cluster - Ward's Method & UnStandardized).
2) Determine No. Of Clusters
3) Use means of Variables to 'name' cluster
4) Run KMeans for the chosen number of clusters
5) Create cross tab of Kmeans variable Vs Wards method to determine how many values have 'moved' clusters
5) Repeat all for subset

I've completed the SPSS word and now I have to write up an executive summary on my analysis of the data...but this is where I am stuck. How should I compare the means data I got from the Hierarchial cluster and the K Means Cluster...how are they related. Generally the values I have for both are very similar (around 0.10 difference) - so what does this mean, are the clusters similar?  Was my original cluster analysis successful?

Does anyone have any advice on how to Anaylse all the SPSS output on SPSS I have..and how could I compare the data from Hierarchial and K Means?

Any help would be greatly appreciated.

Thanks

Confused SPSS User!

[Melissa Ives]

Hey confused SPSS user,

We created this document for Cluster analysis that has information on comparing clusters/selecting the optimal number of clusters.  I suspect you could use something similar to compare Hierarchical vs. K means--although you may want to consider which measure (Hierarchical cluster) theoretically fits your data better.

http://www.chestnut.org/LI/downloads/training_memos/Cluster.pdf

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of mess2010
Sent: Wednesday, December 01, 2010 11:43 AM
To: [hidden email]
Subject: [SPSSX-L] Comparing Output from Hierarchial Clustering Vs K Means

We've been set a task to use Cluster analyse a large quantity of data using PASW(SPSS). The process we followed was:
1) Run Cluster anaylsis (using Hierarchial Cluster - Ward's Method & UnStandardized).
2) Determine No. Of Clusters
3) Use means of Variables to 'name' cluster
4) Run KMeans for the chosen number of clusters
5) Create cross tab of Kmeans variable Vs Wards method to determine how many values have 'moved' clusters
5) Repeat all for subset

I've completed the SPSS word and now I have to write up an executive summary on my analysis of the data...but this is where I am stuck. How should I compare the means data I got from the Hierarchial cluster and the K Means Cluster...how are they related. Generally the values I have for both are very similar (around 0.10 difference) - so what does this mean, are the clusters similar?  Was my original cluster analysis successful?

Does anyone have any advice on how to Anaylse all the SPSS output on SPSS I have..and how could I compare the data from Hierarchial and K Means?

Any help would be greatly appreciated.

Thanks

Confused SPSS User!
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Comparing-Output-from-Hierarchial-Clustering-Vs-K-Means-tp3288054p3288054.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

PRIVILEGED AND CONFIDENTIAL INFORMATION
This transmittal and any attachments may contain PRIVILEGED AND
CONFIDENTIAL information and is intended only for the use of the
addressee. If you are not the designated recipient, or an employee
or agent authorized to deliver such transmittals to the designated
recipient, you are hereby notified that any dissemination,
copying or publication of this transmittal is strictly prohibited. If
you have received this transmittal in error, please notify us
immediately by replying to the sender and delete this copy from your
system. You may also call us at [phone] for assistance.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD